CN117541367A - Digital bidding document making and evaluating system based on artificial intelligence - Google Patents

Digital bidding document making and evaluating system based on artificial intelligence Download PDF

Info

Publication number
CN117541367A
CN117541367A CN202410024751.0A CN202410024751A CN117541367A CN 117541367 A CN117541367 A CN 117541367A CN 202410024751 A CN202410024751 A CN 202410024751A CN 117541367 A CN117541367 A CN 117541367A
Authority
CN
China
Prior art keywords
bidding
file
bid
document
correction
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202410024751.0A
Other languages
Chinese (zh)
Other versions
CN117541367B (en
Inventor
张汪洋
李志强
景莉婷
杨旭
李宇超
周健
于家欢
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Liaoning Netcom Digital Technology Industry Co ltd
Original Assignee
Liaoning Netcom Digital Technology Industry Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Liaoning Netcom Digital Technology Industry Co ltd filed Critical Liaoning Netcom Digital Technology Industry Co ltd
Priority to CN202410024751.0A priority Critical patent/CN117541367B/en
Publication of CN117541367A publication Critical patent/CN117541367A/en
Application granted granted Critical
Publication of CN117541367B publication Critical patent/CN117541367B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L63/00—Network architectures or network communication protocols for network security
    • H04L63/04—Network architectures or network communication protocols for network security for providing a confidential data exchange among entities communicating through data packet networks
    • H04L63/0428—Network architectures or network communication protocols for network security for providing a confidential data exchange among entities communicating through data packet networks wherein the data content is protected, e.g. by encrypting or encapsulating the payload
    • H04L63/0442—Network architectures or network communication protocols for network security for providing a confidential data exchange among entities communicating through data packet networks wherein the data content is protected, e.g. by encrypting or encapsulating the payload wherein the sending and receiving network entities apply asymmetric encryption, i.e. different keys for encryption and decryption
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00—Handling natural language data
    • G06F40/20—Natural language analysis
    • G06F40/253—Grammatical analysis; Style critique
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/10—Office automation; Time management
    • G06Q10/103—Workflow collaboration or project management
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/06—Buying, selling or leasing transactions
    • G06Q30/08—Auctions
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L63/00—Network architectures or network communication protocols for network security
    • H04L63/14—Network architectures or network communication protocols for network security for detecting or protecting against malicious traffic
    • H04L63/1433—Vulnerability analysis
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/01—Protocols
    • H04L67/06—Protocols specially adapted for file transfer, e.g. file transfer protocol [FTP]
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L9/00—Cryptographic mechanisms or cryptographic arrangements for secret or secure communications; Network security protocols
    • H04L9/40—Network security protocols

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Computer Security & Cryptography (AREA)
  • Strategic Management (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Human Resources & Organizations (AREA)
  • Entrepreneurship & Innovation (AREA)
  • General Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Business, Economics & Management (AREA)
  • Computing Systems (AREA)
  • Computer Hardware Design (AREA)
  • Economics (AREA)
  • Marketing (AREA)
  • Finance (AREA)
  • Accounting & Taxation (AREA)
  • Operations Research (AREA)
  • Data Mining & Analysis (AREA)
  • Development Economics (AREA)
  • Quality & Reliability (AREA)
  • Tourism & Hospitality (AREA)
  • Health & Medical Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Computational Linguistics (AREA)
  • General Health & Medical Sciences (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention belongs to the field of digital bidding documents, and particularly discloses a digital bidding document making and evaluating system based on artificial intelligence, which comprises the following steps: the system comprises a case library module, a file generation to be checked module, a collaborative checking module, an evaluation error correction module and a server. The scheme adopts the main body library to store enterprise information, synchronizes the enterprise information to the bidding document, and reduces deviation of manual operation; a method based on case reasoning is adopted to search and reuse historical bidding files to manufacture new bidding files, so that efficiency and applicability are improved; the cloud cooperation architecture is adopted, so that simultaneous editing of multiple persons is realized, and the working efficiency is improved; adopting a novel cloud encryption gateway to control authority of an inspector, and avoiding illegal user access or malicious operation; and (3) checking grammar problems in the bidding documents by using an LM checking model, correcting by using an antagonistic network model, and improving the quality of the bidding documents.

Description

Digital bidding document making and evaluating system based on artificial intelligence
Technical Field
The invention belongs to the field of digital bidding documents, and particularly relates to a digital bidding document making and evaluating system based on artificial intelligence.
Background
The system for making and evaluating the digitized bidding documents based on the artificial intelligence is a system for quickly making the bidding documents by utilizing an artificial intelligence algorithm and boosting the digitized transformation of engineering projects. However, the conventional bidding document production system has the technical problems that enterprise information needs to be manually input, which is time-consuming and is easy to make mistakes; the method has the technical problems that the production of the bidding documents still needs to be carried out manually, the workload is large, and the efficiency is low; the technical problems that bidding documents cannot be simultaneously cooperated when a plurality of examination personnel are required to finish, personal thinking is easy to fall into, and efficiency is low exist; the technical problems that the authority control of users and bidding documents is weaker, sensitive data is easy to leak, and potential security holes exist; there is a technical problem that a lot of time and labor are consumed to manually check and correct errors in the bidding documents, and the errors are easily affected by subjective judgment of individuals.
Disclosure of Invention
Aiming at the defects of the prior art, the invention provides a digital bidding document making and evaluating system based on artificial intelligence, and aiming at the technical problems that enterprise information is required to be manually input and is time-consuming and error-prone, the invention adopts a newly built main body library to store the enterprise information, automatically synchronizes the enterprise information to a document to be inspected, reduces errors and deviations of manual operation, ensures the accuracy and consistency of the information, and provides better data management and information security; aiming at the technical problems that the production of the bidding documents still needs to be carried out manually, the workload is high and the efficiency is low, the method based on case reasoning is adopted to search and reuse the historical bidding documents to formulate the bidding documents, so that the efficiency is higher, and the method is better applicable to complex and changeable actual conditions; aiming at the technical problems that a plurality of examination personnel cannot cooperate at the same time when the bidding documents are completed, personal thinking is easy to fall into and the efficiency is low, a cloud cooperation framework is adopted, so that the bidding documents can be examined and edited by a plurality of persons at the same time, skills and experiences can be shared, the working efficiency and the resource utilization rate are improved, and the diversity and the creativity of the documents are increased; aiming at the technical problems that the authority control of users and bidding documents is weaker, sensitive data is easy to leak and potential security holes exist, a cloud encryption gateway is adopted to control the authority of an inspector, so that illegal users can be prevented from accessing sensitive information or carrying out malicious operation, only authorized personnel can be ensured to edit the bidding documents, and the security of bidding document information is further improved; aiming at the technical problems that errors in the bidding documents need to be checked and corrected manually, a great deal of time and labor are consumed, and the problems are easily influenced by personal subjective judgment, an LM checking model is adopted to check grammar problems in the bidding documents, and an antagonistic network model is adopted to correct grammar problems in the bidding documents.
The technical scheme adopted by the invention is as follows: the invention provides an artificial intelligence-based digital bidding document production and evaluation system, which comprises a case library creation module, a document to be checked generation module, a collaborative checking module, an evaluation error correction module and a server, wherein the case library creation module is used for creating a case library;
the case library creating module is used for collecting historical bidding documents and transmitting the historical bidding documents and the bidding documents to the case library;
the file generation to be checked module adopts a method based on case reasoning, searches and reuses historical bidding files according to the requirement of bidding files, and makes the file to be checked;
the collaboration checking module realizes authority control of all checking staff editing the files to be checked on line through cloud collaboration and a cloud encryption gateway, and the checking staff edits and checks the files to be checked to generate an intermediate bidding file;
the evaluation error correction module adopts an LM check model to check grammar problems in the middle bidding file, and uses an antagonistic network model to correct grammar problems in the bidding file to manufacture a final bidding file;
the server is provided with a main body library, the main body library stores certificate information of bidding companies, including enterprise business licenses, tax registration certificates, organization code certificates, legal representative personal identification certificates and qualification certificates, and generates a responsible person public key and a responsible person private key for bidding responsible persons and generates a agency public key and a agency private key for censoring persons;
the digital bidding document making and evaluating system based on artificial intelligence adopts a digital bidding document making and evaluating method to make and evaluate the bidding document;
the digital bidding document making and evaluating method comprises the following steps:
step S1: creating a case library, wherein the case library is stored in a server, a historical bid file and a bid-tendering file are acquired, the historical bid file and the bid-tendering file are transmitted to the case library, the historical bid file comprises a project type, a purchasing method, whether prequalification, a bid evaluation method, a bid object type, enterprise qualification and a technical scheme, the minimum unit of the technical scheme is a technical point, the technical point comprises an attribute value and a specific solution, and the bid-tendering file comprises a project type, a purchasing method, whether prequalification, a bid evaluation method and a bid object type;
step S2: generating a file to be checked, wherein the file to be checked comprises a technical scheme and enterprise qualification, extracting historical bidding files from a case library to form a bidding file similar set according to similarity, constructing any non-interference sequence, calculating a non-interference sequence index of a technical point of each historical bidding file in the bidding file similar set, synthesizing bidding templates to form a bidding template set according to the non-interference sequence index, calculating a recommended value of the bidding templates, selecting a bidding template with the highest recommended value to generate the file to be checked, and transmitting the file to be checked to a collaborative review module;
step S3: collaborative inspection, receiving a file to be inspected, realizing authority control of all inspection personnel editing the file to be inspected on line through cloud collaboration and a cloud encryption gateway, editing and inspecting the file to be inspected by the inspection personnel, generating a middle bidding file, and transmitting the middle bidding file to an evaluation error correction module;
step S4: evaluating and correcting errors, receiving the middle bidding file, checking grammar problems in the middle bidding file by using an LM checking model, identifying sentences to be corrected, correcting the sentences to be corrected by using a resistance network model, and manufacturing a final bidding file.
Further, in step S2, a file to be examined is generated, which specifically includes the following steps:
step S21: defining bidding case characteristics, wherein the bidding case characteristics comprise item types, purchasing methods, whether prequalification is carried out, bid evaluation methods and bidding object types;
step S22: the dissimilarity coefficient of the bid case characteristics of the bid documents and the bid case characteristics of each historical bid document are calculated using the following formula:
；
in the method, in the process of the invention,is a historical bidding document in the case library,is a bidding document which is used for the bidding,is the total number of bid case features,=5，is the first bidding case featureThe characteristics of the device are that,is thatAndis used to determine the degree of non-similarity of the coefficients of (c),is thatIs the first of (2)The values of the characteristics of the individual bid cases,is thatIs the first of (2)The values of the characteristics of the individual bid cases,is thatFrom 1 toIs the sum of the sums of the (3);
step S23: the similarity of the bid case characteristics of the bid documents and the bid case characteristics of each historical bid document is calculated by the following formula:
；
in the method, in the process of the invention,is thatAndis used for the degree of similarity of (c) to (c),is satisfied withIs defined by the number of features of (a),is satisfied withThe number of features of (2);
step S24: putting the historical bidding documents with similarity greater than 0.7 into a bidding document similar set, and outputting the bidding document similar set;
step S25: randomly generating a set of sequences satisfying the first thereofItems greater than beforeThe condition of the sum of the terms, the sequence being noted as a non-interfering sequence；
Step S26: the non-interference sequence index of each technical point of the historical bidding documents in the similar set of bidding documents is calculated, and the following formula is used:
；
in the method, in the process of the invention,is the technical point of each historical bidding document in the similar collection of bidding documentsThe non-interfering sequence index(s) is (are) used,is the first of the historical bidding documentsIn the light of the above-mentioned technical points,is thatIs used to determine the value of the attribute of (c),is thatIs the first of (2)The sequence of items is selected from the list,is thatIs used to determine the total number of sequences of (a),is thatFrom 1 toIs the sum of the sums of the (3);
step S27: combining the specific solutions of the technical points with the same non-interference sequence index into one bidding template, wherein all the bidding templates form a bidding template set;
step S28: the recommended value of each bid template in the bid template set is calculated by the following formula:
；
in the method, in the process of the invention,is the firstThe number of bid templates is selected to be the number of bid templates,is the firstThe recommended value of each bid template is calculated,is thatThe frequency of occurrence in the set of bid templates,is the total number of historical bid files in the similar collection of bid files;
step S29: generating a file to be inspected, arranging recommended values of bid templates in a bid template set in descending order, taking the bid template with the largest recommended value as a technical scheme of the file to be inspected, and automatically synchronizing certificate information of a bidding company from a main body library by a server to serve as enterprise qualification of the file to be inspected;
further, in step S3, collaborative review specifically includes the steps of:
step S31: the bidding responsible person generates a responsible person public key and a responsible person private key through a server, generates a safe and unique file key for a file to be checked, generates a proxy public key and a proxy private key for a checking person, creates a permission object which indicates that the checking person can access the file to be checked, and stores the responsible person public key, the proxy public key and the permission object in the server;
step S32: the bidding responsible party uses the file key to carry out AES-256 encryption on the file to be inspected to generate an encrypted file;
step S33: the bidding responsible party encrypts a file key by using a responsible party public key, generates an encrypted file key P, and stores the encrypted file key P at the beginning of an encrypted file;
step S34: when an inspector accesses a file to be inspected, an encrypted file key P is extracted from the beginning of the encrypted file;
step S35: sending the encrypted file key P to the server by the inspector for re-encryption;
step S36: the server checks whether the censor has permission objects, judges whether the censor is granted permission to open the file to be censored, if not, the server refuses the censor to access the file to be censored, otherwise, the server decrypts the encrypted file key P by using the public key of the responsible person, re-encrypts the encrypted file key P by using the proxy public key of the censor, generates a file key ciphertext, and sends the file key ciphertext to the censor;
step S37: the censor decrypts the file key ciphertext through the own proxy private key to obtain the file key, and decrypts the encrypted file by using the file key, so that the file to be censored can be accessed;
step S38: editing and inspecting the file to be inspected by an inspector to generate an intermediate bidding file;
further, in step S4, the error correction is evaluated, specifically including the steps of:
step S41: the grammar error check specifically comprises the following steps:
step S411: collecting a common vocabulary dictionary, wherein the common vocabulary dictionary comprises common vocabularies and synonyms thereof in a bidding document;
step S412: dividing the middle bidding document into original sentences, matching the original sentences with words in a common word dictionary to divide words to obtain word sets, replacing, deleting or inserting the words in the word sets by using synonyms, adjusting the positions of noun phrases in the original sentences, changing the positions of vergence, generating disturbance sentences, and forming a near neighborhood by all the disturbance sentences and the original sentences;
step S413: calculating the distribution probability of the disturbance sentences and the original sentences in the neighboring domain by using the LM model;
step S414: is provided withThe distribution probability threshold value isIf the distribution probability of the original sentence is higher thanThe original sentence is considered to be a correct sentence, otherwise the original sentence is a sentence to be corrected;
step S415: when the to-be-corrected statement exists in the middle bidding document, executing step S42, otherwise, acquiring the item type, the purchasing method, whether to conduct qualification pre-examination, the bid evaluation method and the bidding object type from the bidding document, merging the item type, the purchasing method and the bidding object type with the middle bidding document to form a final bidding document, outputting the final bidding document, and adding the final bidding document as a new historical bidding document into the case library;
step S42: the grammar error correction specifically comprises the following steps:
step S421: collecting a correction data set, wherein the correction data set consists of pairs of error sentences and corresponding correction sentences;
step S422: establishing and initializing an antagonistic network model, wherein the antagonistic network model comprises a generator and a discriminator;
step S423: inputting the correction data set into a generator for MLE pre-training;
step S424: the generator generates correction sentences for error sentences of the correction data set, and all pairs of the error sentences and the correction sentences form a correction pair set;
step S425: inputting the correction data set and the correction pair set into a discriminator for BCE pre-training;
step S426: the generator and the discriminator perform resistance learning, the generator generates a correction sentence and transmits the correction sentence to the discriminator, the discriminator calculates the accuracy of the correction sentence, when the accuracy is smaller than 0.4, the accuracy is fed back to the generator as a reward, the generator is updated by a Monte Carlo strategy gradient method, and the Monte Carlo strategy gradient is calculated by the following formula:
；
；
in the method, in the process of the invention,is the update gradient of the arbiter and,is the update gradient of the generator and,andthe initial weights of the generator and the arbiter respectively,is an error statement that is to be read,is a correction statement in the correction dataset,is a correction sentence generated by the generator;
step S427: when the accuracy is greater than 0.4, the correction sentence is used as a correction sentence, the generator performs Teacher-mapping learning, and step S426 is repeatedly executed until the error sentence traversal is completed, and the resistance learning is finished;
step S428: the sentence to be corrected is input into the antagonistic network model, the corrected sentence is output to the antagonistic network model, and the sentence to be corrected is replaced with the corrected sentence, and step S41 is repeatedly executed.
The beneficial results obtained by adopting the scheme of the invention are as follows:
(1) Aiming at the technical problems that the manual input of enterprise information is needed, time is consumed and errors are easy to occur, the main body library is adopted to store the enterprise information, the enterprise information is automatically synchronized to the file to be checked, errors and deviations of manual operation are reduced, the accuracy and consistency of the information are ensured, and better data management and information security are provided;
(2) Aiming at the technical problems that the production of the bidding documents still needs to be carried out manually, the workload is high and the efficiency is low, the method adopts a case-based reasoning method to obtain the documents to be examined, searches and reuses the historical bidding documents to formulate new bidding documents, and not only has higher efficiency, but also can be better suitable for complex and changeable actual conditions;
(3) Aiming at the technical problems that a plurality of examination staff cannot cooperate at the same time when the bidding documents are completed, and are easy to fall into personal thinking and low in efficiency, the cloud cooperation framework is adopted, so that the bidding documents can be examined and edited by a plurality of persons at the same time, skills and experiences can be shared, the working efficiency and the resource utilization rate are improved, and the diversity and creativity of the documents are improved;
(4) Aiming at the technical problems that sensitive data are easy to leak and potential security holes are caused due to weaker authority control on users and bidding files, the novel cloud encryption gateway is adopted to protect public cloud storage, authority control is carried out on modification of the bidding files, illegal users can be prevented from accessing sensitive information or carrying out malicious operation, only authorized users can edit the bidding files, and the security of bidding file information is further improved;
(5) Aiming at the technical problems that errors in the bidding documents need to be checked and corrected manually, a great deal of time and labor are consumed and the problems are easily influenced by personal subjective judgment, the LM checking model is adopted to check grammar problems in the bidding documents, and the antagonistic network model is used to correct grammar problems in the bidding documents, so that the quality of the bidding documents is improved.
Drawings
FIG. 1 is a schematic diagram of an artificial intelligence based digitized bid document production and evaluation system;
fig. 2 is a flow chart of a method based on case-based reasoning.
The accompanying drawings are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate the invention and together with the embodiments of the invention, serve to explain the invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are only some, but not all embodiments of the invention; all other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
In the description of the present invention, it should be understood that the terms "upper," "lower," "front," "rear," "left," "right," "top," "bottom," "inner," "outer," and the like indicate orientation or positional relationships based on those shown in the drawings, merely to facilitate description of the invention and simplify the description, and do not indicate or imply that the devices or elements referred to must have a particular orientation, be constructed and operated in a particular orientation, and thus should not be construed as limiting the invention.
Embodiment one: referring to fig. 1, the invention provides an artificial intelligence based digital bidding document production and evaluation system, which comprises a case library creation module, a document to be checked generation module, a collaborative inspection module, an evaluation error correction module and a server;
the case library creating module is used for collecting historical bidding documents and transmitting the historical bidding documents and the bidding documents to the case library;
the file generation to be checked module adopts a method based on case reasoning, searches and reuses historical bidding files according to the requirement of bidding files, and makes the file to be checked;
the collaboration checking module realizes authority control of all checking staff editing the files to be checked on line through cloud collaboration and a cloud encryption gateway, and the checking staff edits and checks the files to be checked to generate an intermediate bidding file;
the evaluation error correction module adopts an LM check model to check grammar problems in the middle bidding file, and uses an antagonistic network model to correct grammar problems in the bidding file to manufacture a final bidding file;
the server is provided with a main body library, the main body library stores certificate information of bidding companies, including enterprise business licenses, tax registration certificates, organization code certificates, legal representative personal identification certificates and qualification certificates, and generates a responsible person public key and a responsible person private key for bidding responsible persons and generates a agency public key and a agency private key for censoring persons;
the digital bidding document making and evaluating system based on artificial intelligence adopts a digital bidding document making and evaluating method to make and evaluate the bidding document;
the digital bidding document making and evaluating method comprises the following steps:
step S1: creating a case library, wherein the case library is stored in a server, a historical bid file and a bid-tendering file are acquired, the historical bid file and the bid-tendering file are transmitted to the case library, the historical bid file comprises a project type, a purchasing method, whether prequalification, a bid evaluation method, a bid object type, enterprise qualification and a technical scheme, the minimum unit of the technical scheme is a technical point, the technical point comprises an attribute value and a specific solution, and the bid-tendering file comprises a project type, a purchasing method, whether prequalification, a bid evaluation method and a bid object type;
step S2: generating a file to be checked, wherein the file to be checked comprises a technical scheme and enterprise qualification, extracting historical bidding files from a case library to form a bidding file similar set according to similarity, constructing any non-interference sequence, calculating a non-interference sequence index of a technical point of each historical bidding file in the bidding file similar set, synthesizing bidding templates to form a bidding template set according to the non-interference sequence index, calculating a recommended value of the bidding templates, selecting a bidding template with the highest recommended value to generate the file to be checked, and transmitting the file to be checked to a collaborative review module;
step S3: collaborative inspection, receiving a file to be inspected, realizing authority control of all inspection personnel editing the file to be inspected on line through cloud collaboration and a cloud encryption gateway, editing and inspecting the file to be inspected by the inspection personnel, generating a middle bidding file, and transmitting the middle bidding file to an evaluation error correction module;
step S4: evaluating and correcting errors, receiving the middle bidding file, checking grammar problems in the middle bidding file by using an LM checking model, identifying sentences to be corrected, correcting the sentences to be corrected by using a resistance network model, and manufacturing a final bidding file.
Embodiment two: referring to fig. 1 and 2, the embodiment is based on the above embodiment, and in step S2, the file to be examined is generated, which specifically includes the following steps:
step S21: defining bidding case characteristics, wherein the bidding case characteristics comprise item types, purchasing methods, whether prequalification is carried out, bid evaluation methods and bidding object types;
step S22: the dissimilarity coefficient of the bid case characteristics of the bid documents and the bid case characteristics of each historical bid document are calculated using the following formula:
；
in the method, in the process of the invention,is a historical bidding document in the case library,is a bidding document which is used for the bidding,is the total number of bid case features,=5，is the first bidding case featureThe characteristics of the device are that,is thatAndis used to determine the degree of non-similarity of the coefficients of (c),is thatIs the first of (2)The values of the characteristics of the individual bid cases,is thatIs the first of (2)The values of the characteristics of the individual bid cases,is thatFrom 1 toIs the sum of the sums of the (3);
step S23: the similarity of the bid case characteristics of the bid documents and the bid case characteristics of each historical bid document is calculated by the following formula:
；
in the method, in the process of the invention,is thatAndis used for the degree of similarity of (c) to (c),is satisfied withIs defined by the number of features of (a),is satisfied withThe number of features of (2);
step S24: putting the historical bidding documents with similarity greater than 0.7 into a bidding document similar set, and outputting the bidding document similar set;
step S25: randomly generating a set of sequences satisfying the first thereofItems greater than beforeThe condition of the sum of the terms, the sequence being noted as a non-interfering sequence；
Step S26: the non-interference sequence index of each technical point of the historical bidding documents in the similar set of bidding documents is calculated, and the following formula is used:
；
in the method, in the process of the invention,is a non-interference sequence index of the technical points of each historical bidding document in the similar set of bidding documents,is the first of the historical bidding documentsIn the light of the above-mentioned technical points,is thatIs used to determine the value of the attribute of (c),is thatIs the first of (2)The sequence of items is selected from the list,is thatIs used to determine the total number of sequences of (a),is thatFrom 1 toIs the sum of the sums of the (3);
step S27: combining the specific solutions of the technical points with the same non-interference sequence index into one bidding template, wherein all the bidding templates form a bidding template set;
step S28: the recommended value of each bid template in the bid template set is calculated by the following formula:
；
in the method, in the process of the invention,is the firstThe number of bid templates is selected to be the number of bid templates,is the firstThe recommended value of each bid template is calculated,is thatThe frequency of occurrence in the set of bid templates,is the total number of historical bid files in the similar collection of bid files;
step S29: generating a file to be inspected, arranging recommended values of bid templates in a bid template set in descending order, taking the bid template with the largest recommended value as a technical scheme of the file to be inspected, and automatically synchronizing certificate information of a bidding company from a main body library by a server to serve as enterprise qualification of the file to be inspected.
Through the operation, aiming at the technical problems that the production of the bidding documents still needs to be carried out manually, and the workload is high and the efficiency is low, the method adopts a case-based reasoning method to acquire the documents to be examined, searches and reuses the historical bidding documents to formulate new bidding documents, so that the efficiency is higher, and the method is better applicable to complex and changeable actual conditions.
Embodiment III: referring to fig. 1 and 2, the embodiment is based on the above embodiment, and in step S3, the collaborative review specifically includes the following steps:
step S31: the bidding responsible person generates a responsible person public key and a responsible person private key through a server, generates a safe and unique file key for a file to be checked, generates a proxy public key and a proxy private key for a checking person, creates a permission object which indicates that the checking person can access the file to be checked, and stores the responsible person public key, the proxy public key and the permission object in the server;
step S32: the bidding responsible party uses the file key to carry out AES-256 encryption on the file to be inspected to generate an encrypted file;
step S33: the bidding responsible party encrypts a file key by using a responsible party public key, generates an encrypted file key P, and stores the encrypted file key P at the beginning of an encrypted file;
step S34: when an inspector accesses a file to be inspected, an encrypted file key P is extracted from the beginning of the encrypted file;
step S35: sending the encrypted file key P to the server by the inspector for re-encryption;
step S36: the server checks whether the censor has permission objects, judges whether the censor is granted permission to open the file to be censored, if not, the server refuses the censor to access the file to be censored, otherwise, the server decrypts the encrypted file key P by using the public key of the responsible person, re-encrypts the encrypted file key P by using the proxy public key of the censor, generates a file key ciphertext, and sends the file key ciphertext to the censor;
step S37: the censor decrypts the file key ciphertext through the own proxy private key to obtain the file key, and decrypts the encrypted file by using the file key, so that the file to be censored can be accessed;
step S38: the censoring personnel edits and censores the document to be censored, and generates an intermediate bidding document.
Through the operation, aiming at the technical problems that when the bidding documents are finished by a plurality of examination personnel, the bidding documents cannot be simultaneously cooperated, personal thinking is easy to fall into, and the efficiency is low, a cloud cooperation framework is adopted, so that the simultaneous examination and editing of the bidding documents by a plurality of persons are realized, skills and experiences can be shared, the working efficiency and the resource utilization rate are improved, and the diversity and the creativity of the documents are increased; aiming at the technical problems that users and bidding documents are weaker in authority control, sensitive data are easy to leak, and potential security holes exist, a novel cloud encryption gateway is adopted to protect public cloud storage, authority control is carried out on modification of the bidding documents, illegal users can be prevented from accessing sensitive information or carrying out malicious operation, only authorized users are ensured to edit the bidding documents, and the security of bidding document information is further improved.
Embodiment four: referring to fig. 1 and 2, the embodiment is based on the above embodiment, and in step S4, the error correction is evaluated, specifically including the steps of:
step S41: the grammar error check specifically comprises the following steps:
step S411: collecting a common vocabulary dictionary, wherein the common vocabulary dictionary comprises common vocabularies and synonyms thereof in a bidding document;
step S412: dividing the middle bidding document into original sentences, matching the original sentences with words in a common word dictionary to divide words to obtain word sets, replacing, deleting or inserting the words in the word sets by using synonyms, adjusting the positions of noun phrases in the original sentences, changing the positions of vergence, generating disturbance sentences, and forming a near neighborhood by all the disturbance sentences and the original sentences;
step S413: calculating the distribution probability of the disturbance sentences and the original sentences in the neighboring domain by using the LM model;
step S414: let the distribution probability threshold beIf the distribution probability of the original sentence is higher thanThe original sentence is considered to be a correct sentence, otherwise the original sentence is a sentence to be corrected;
step S415: when the to-be-corrected statement exists in the middle bidding document, executing step S42, otherwise, acquiring the item type, the purchasing method, whether to conduct qualification pre-examination, the bid evaluation method and the bidding object type from the bidding document, merging the item type, the purchasing method and the bidding object type with the middle bidding document to form a final bidding document, outputting the final bidding document, and adding the final bidding document as a new historical bidding document into the case library;
step S42: the grammar error correction specifically comprises the following steps:
step S421: collecting a correction data set, wherein the correction data set consists of pairs of error sentences and corresponding correction sentences;
step S422: establishing and initializing an antagonistic network model, wherein the antagonistic network model comprises a generator and a discriminator;
step S423: inputting the correction data set into a generator for MLE pre-training;
step S424: the generator generates correction sentences for error sentences of the correction data set, and all pairs of the error sentences and the correction sentences form a correction pair set;
step S425: inputting the correction data set and the correction pair set into a discriminator for BCE pre-training;
step S426: the generator and the discriminator perform resistance learning, the generator generates a correction sentence and transmits the correction sentence to the discriminator, the discriminator calculates the accuracy of the correction sentence, when the accuracy is smaller than 0.4, the accuracy is fed back to the generator as a reward, the generator is updated by a Monte Carlo strategy gradient method, and the Monte Carlo strategy gradient is calculated by the following formula:
；
；
in the method, in the process of the invention,is the update gradient of the arbiter and,is the update gradient of the generator and,andthe initial weights of the generator and the arbiter respectively,is an error statement that is to be read,is a correction statement in the correction dataset,is a correction sentence generated by the generator;
step S427: when the accuracy is greater than 0.4, the correction sentence is used as a correction sentence, the generator performs Teacher-mapping learning, and step S426 is repeatedly executed until the error sentence traversal is completed, and the resistance learning is finished;
step S428: the sentence to be corrected is input into the antagonistic network model, the corrected sentence is output to the antagonistic network model, and the sentence to be corrected is replaced with the corrected sentence, and step S41 is repeatedly executed.
Through the operation, aiming at the technical problems that the enterprise information is required to be manually input, time is consumed and errors are easy to occur, the enterprise information is stored by adopting the main body library, the enterprise information is automatically synchronized to the file to be checked, errors and deviation of manual operation are reduced, the accuracy and consistency of the information are ensured, and better data management and information security are provided; aiming at the technical problems that errors in the bidding documents need to be checked and corrected manually, a great deal of time and labor are consumed and the problems are easily influenced by personal subjective judgment, the LM checking model is adopted to check grammar problems in the bidding documents, and the antagonistic network model is used to correct grammar problems in the bidding documents, so that the quality of the bidding documents is improved.
It is noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus.
Although embodiments of the present invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made therein without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.
The invention and its embodiments have been described above with no limitation, and the actual construction is not limited to the embodiments of the invention as shown in the drawings. In summary, if one of ordinary skill in the art is informed by this disclosure, a structural manner and an embodiment similar to the technical solution should not be creatively devised without departing from the gist of the present invention.

Claims (5)

1. The digital bidding document making and evaluating system based on the artificial intelligence is characterized by comprising a case library creating module, a document to be inspected generating module, a cooperation inspection module, an evaluation error correction module and a server;
the case library creating module is used for collecting historical bidding documents and transmitting the historical bidding documents and the bidding documents to the case library;
the file generation to be checked module adopts a method based on case reasoning, searches and reuses historical bidding files according to the requirement of bidding files, and makes the file to be checked;
the collaboration checking module realizes authority control of all checking staff editing the files to be checked on line through cloud collaboration and a cloud encryption gateway, and the checking staff edits and checks the files to be checked to generate an intermediate bidding file;
the evaluation error correction module adopts an LM check model to check grammar problems in the middle bidding file, and uses an antagonistic network model to correct grammar problems in the bidding file to manufacture a final bidding file;
the server is provided with a main body library, the main body library stores certificate information of bidding companies, including business licenses, tax registration certificates, organization code certificates, legal representatives' identity certificates and qualification certificates, and generates a responsible person public key and a responsible person private key for bidding responsible persons and generates a agency public key and a agency private key for censoring persons.
2. The digital bid document creation and evaluation system based on artificial intelligence according to claim 1, wherein the digital bid document creation and evaluation system based on artificial intelligence is implemented by using a digital bid document creation and evaluation method, and the digital bid document creation and evaluation method specifically comprises the following steps:
step S1: creating a case library, wherein the case library is stored in a server, a historical bid file and a bid-tendering file are acquired, the historical bid file and the bid-tendering file are transmitted to the case library, the historical bid file comprises a project type, a purchasing method, whether prequalification, a bid evaluation method, a bid object type, enterprise qualification and a technical scheme, the minimum unit of the technical scheme is a technical point, the technical point comprises an attribute value and a specific solution, and the bid-tendering file comprises a project type, a purchasing method, whether prequalification, a bid evaluation method and a bid object type;
step S2: generating a file to be checked, wherein the file to be checked comprises a technical scheme and enterprise qualification, extracting historical bidding files from a case library to form a bidding file similar set according to similarity, constructing any non-interference sequence, calculating a non-interference sequence index of a technical point of each historical bidding file in the bidding file similar set, synthesizing bidding templates to form a bidding template set according to the non-interference sequence index, calculating a recommended value of the bidding templates, selecting a bidding template with the highest recommended value to generate the file to be checked, and transmitting the file to be checked to a collaborative review module;
step S3: collaborative inspection, receiving a file to be inspected, realizing authority control of all inspection personnel editing the file to be inspected on line through cloud collaboration and a cloud encryption gateway, editing and inspecting the file to be inspected by the inspection personnel, generating a middle bidding file, and transmitting the middle bidding file to an evaluation error correction module;
step S4: evaluating and correcting errors, receiving the middle bidding file, checking grammar problems in the middle bidding file by using an LM checking model, identifying sentences to be corrected, correcting the sentences to be corrected by using a resistance network model, and manufacturing a final bidding file.
3. The digital bid document creation and evaluation system based on artificial intelligence of claim 2, wherein in step S2, the generation of the document to be examined specifically comprises the steps of:
step S21: defining bidding case characteristics, wherein the bidding case characteristics comprise item types, purchasing methods, whether prequalification is carried out, bid evaluation methods and bidding object types;
step S22: the dissimilarity coefficient of the bid case characteristics of the bid documents and the bid case characteristics of each historical bid document are calculated using the following formula:
；
in the method, in the process of the invention,is a historical bidding document in the case base, +.>Is a logo file,/->Is the total number of bidding case features, +.>=5，/>Is the +.f. of the bidding case feature>Personal characteristics (I)>Is->And->Is>Is->Is>The value of the individual bidding case characteristics, +.>Is->Is>The value of the individual bidding case characteristics, +.>Is->From 1 to->Is the sum of the sums of the (3);
step S23: the similarity of the bid case characteristics of the bid documents and the bid case characteristics of each historical bid document is calculated by the following formula:
；
in the method, in the process of the invention,is->And->Similarity of->Is satisfied->The number of features of>Is satisfied->The number of features of (2);
step S24: putting the historical bidding documents with similarity greater than 0.7 into a bidding document similar set, and outputting the bidding document similar set;
step S25: randomly generating a set of sequences satisfying the first thereofThe item is greater than +.>The condition of the sum of the terms, the sequence is marked as a non-interfering sequence->；
Step S26: the non-interference sequence index of each technical point of the historical bidding documents in the similar set of bidding documents is calculated, and the following formula is used:
；
in the method, in the process of the invention,is a non-interference sequence index of technical points of each historical bidding document in the similar set of bidding documents, ++>Is the +.>Technical points (I)>Is->Attribute value of->Is->Is>Item sequence,/->Is->Is used to determine the total number of sequences of (a),is->From 1 to->Is the sum of the sums of the (3);
step S27: combining the specific solutions of the technical points with the same non-interference sequence index into one bidding template, wherein all the bidding templates form a bidding template set;
step S28: the recommended value of each bid template in the bid template set is calculated by the following formula:
；
in the method, in the process of the invention,is->Bid templates->Is->Recommended value of each bid template, +.>Is->Frequency of occurrence in the bid template set, +.>Is the total number of historical bid files in the similar collection of bid files;
step S29: generating a file to be inspected, arranging recommended values of bid templates in a bid template set in descending order, taking the bid template with the largest recommended value as a technical scheme of the file to be inspected, and automatically synchronizing certificate information of a bidding company from a main body library by a server to serve as enterprise qualification of the file to be inspected.
4. The digital bid document production and evaluation system based on artificial intelligence of claim 3, wherein in step S3, the collaborative review specifically comprises the following steps:
step S31: the bidding responsible person generates a responsible person public key and a responsible person private key through a server, generates a safe and unique file key for a file to be checked, generates a proxy public key and a proxy private key for a checking person, creates a permission object which indicates that the checking person can access the file to be checked, and stores the responsible person public key, the proxy public key and the permission object in the server;
step S32: the bidding responsible party uses the file key to carry out AES-256 encryption on the file to be inspected to generate an encrypted file;
step S33: the bidding responsible party encrypts a file key by using a responsible party public key, generates an encrypted file key P, and stores the encrypted file key P at the beginning of an encrypted file;
step S34: when an inspector accesses a file to be inspected, an encrypted file key P is extracted from the beginning of the encrypted file;
step S35: sending the encrypted file key P to the server by the inspector for re-encryption;
step S36: the server checks whether the censor has permission objects, judges whether the censor is granted permission to open the file to be censored, if not, the server refuses the censor to access the file to be censored, otherwise, the server decrypts the encrypted file key P by using the public key of the responsible person, re-encrypts the encrypted file key P by using the proxy public key of the censor, generates a file key ciphertext, and sends the file key ciphertext to the censor;
step S37: the censor decrypts the file key ciphertext through the own proxy private key to obtain the file key, and decrypts the encrypted file by using the file key, so that the file to be censored can be accessed;
step S38: the censoring personnel edits and censores the document to be censored, and generates an intermediate bidding document.
5. The digital bid document production and evaluation system based on artificial intelligence of claim 4, wherein in step S4, the evaluation error correction specifically comprises the steps of:
step S41: the grammar error check specifically comprises the following steps:
step S411: collecting a common vocabulary dictionary, wherein the common vocabulary dictionary comprises common vocabularies and synonyms thereof in a bidding document;
step S412: dividing the middle bidding document into original sentences, matching the original sentences with words in a common word dictionary to divide words to obtain word sets, replacing, deleting or inserting the words in the word sets by using synonyms, adjusting the positions of noun phrases in the original sentences, changing the positions of vergence, generating disturbance sentences, and forming a near neighborhood by all the disturbance sentences and the original sentences;
step S413: calculating the distribution probability of the disturbance sentences and the original sentences in the neighboring domain by using the LM model;
step S414: let the distribution probability threshold beIf the distribution probability of the original sentence is higher than +.>The original sentence is considered to be a correct sentence, otherwise the original sentence is a sentence to be corrected;
step S415: when the to-be-corrected statement exists in the middle bidding document, executing step S42, otherwise, acquiring the item type, the purchasing method, whether to conduct qualification pre-examination, the bid evaluation method and the bidding object type from the bidding document, merging the item type, the purchasing method and the bidding object type with the middle bidding document to form a final bidding document, outputting the final bidding document, and adding the final bidding document as a new historical bidding document into the case library;
step S42: the grammar error correction specifically comprises the following steps:
step S421: collecting a correction data set, wherein the correction data set consists of pairs of error sentences and corresponding correction sentences;
step S422: establishing and initializing an antagonistic network model, wherein the antagonistic network model comprises a generator and a discriminator;
step S423: inputting the correction data set into a generator for MLE pre-training;
step S424: the generator generates correction sentences for error sentences of the correction data set, and all pairs of the error sentences and the correction sentences form a correction pair set;
step S425: inputting the correction data set and the correction pair set into a discriminator for BCE pre-training;
step S426: the generator and the discriminator perform resistance learning, the generator generates a correction sentence and transmits the correction sentence to the discriminator, the discriminator calculates the accuracy of the correction sentence, when the accuracy is smaller than 0.4, the accuracy is fed back to the generator as a reward, the generator is updated by a Monte Carlo strategy gradient method, and the Monte Carlo strategy gradient is calculated by the following formula:
；
；
in the method, in the process of the invention,is the update gradient of the arbiter, +.>Is the update gradient of the generator,>and->Initial weights of generator and arbiter, respectively, +.>Is an error statement +.>Is a correction statement in the correction dataset, +.>Is a correction sentence generated by the generator;
step S427: when the accuracy is greater than 0.4, the correction sentence is used as a correction sentence, the generator performs Teacher-mapping learning, and step S426 is repeatedly executed until the error sentence traversal is completed, and the resistance learning is finished;
step S428: the sentence to be corrected is input into the antagonistic network model, the corrected sentence is output to the antagonistic network model, and the sentence to be corrected is replaced with the corrected sentence, and step S41 is repeatedly executed.
CN202410024751.0A 2024-01-08 2024-01-08 Digital bidding document making and evaluating system based on artificial intelligence Active CN117541367B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202410024751.0A CN117541367B (en) 2024-01-08 2024-01-08 Digital bidding document making and evaluating system based on artificial intelligence

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202410024751.0A CN117541367B (en) 2024-01-08 2024-01-08 Digital bidding document making and evaluating system based on artificial intelligence

Publications (2)

Publication Number Publication Date
CN117541367A true CN117541367A (en) 2024-02-09
CN117541367B CN117541367B (en) 2024-04-02

Family

ID=89794165

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202410024751.0A Active CN117541367B (en) 2024-01-08 2024-01-08 Digital bidding document making and evaluating system based on artificial intelligence

Country Status (1)

Country Link
CN (1) CN117541367B (en)

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110730184A (en) * 2019-10-22 2020-01-24 江苏先安科技有限公司 Novel bidding encryption and decryption method based on SM2 cryptographic algorithm
CN112632943A (en) * 2020-09-30 2021-04-09 中国神华国际工程有限公司 Intelligent bid evaluation method and system, storage medium and electronic device
CN113672972A (en) * 2021-07-01 2021-11-19 国网浙江省电力有限公司建设分公司 Important asset safety monitoring method based on middleboxes
CN114462960A (en) * 2022-01-07 2022-05-10 武汉理工大学 Automatic qualification auditing method and system in electronic bidding
CN114708072A (en) * 2022-03-21 2022-07-05 西安翻译学院 Electronic bidding management system based on big data
CN115694808A (en) * 2022-10-31 2023-02-03 上海晏鼠计算机技术股份有限公司 Cloud collaboration system based on block chain
CN115689696A (en) * 2022-11-03 2023-02-03 安徽皖电招标有限公司 Intelligent bid evaluation method and system based on artificial intelligence technology
CN115760258A (en) * 2022-11-04 2023-03-07 中铁一局集团建筑安装工程有限公司 Intelligent bid document generation method, system, computer device and storage medium

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110730184A (en) * 2019-10-22 2020-01-24 江苏先安科技有限公司 Novel bidding encryption and decryption method based on SM2 cryptographic algorithm
CN112632943A (en) * 2020-09-30 2021-04-09 中国神华国际工程有限公司 Intelligent bid evaluation method and system, storage medium and electronic device
CN113672972A (en) * 2021-07-01 2021-11-19 国网浙江省电力有限公司建设分公司 Important asset safety monitoring method based on middleboxes
CN114462960A (en) * 2022-01-07 2022-05-10 武汉理工大学 Automatic qualification auditing method and system in electronic bidding
CN114708072A (en) * 2022-03-21 2022-07-05 西安翻译学院 Electronic bidding management system based on big data
CN115694808A (en) * 2022-10-31 2023-02-03 上海晏鼠计算机技术股份有限公司 Cloud collaboration system based on block chain
CN115689696A (en) * 2022-11-03 2023-02-03 安徽皖电招标有限公司 Intelligent bid evaluation method and system based on artificial intelligence technology
CN115760258A (en) * 2022-11-04 2023-03-07 中铁一局集团建筑安装工程有限公司 Intelligent bid document generation method, system, computer device and storage medium

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
K.WANG等: ""Automatic generation of sentimental texts via mixture adversarial networks"", 《ARTIFICIAL INTELLIGENCE》, vol. 275, 19 July 2019 (2019-07-19) *
T.T. ZHOU等: ""Construction Method of Tender Document Based on Case-based Reasoning"", 《CCC PUBLICATIONS》, vol. 16, no. 3, 30 June 2021 (2021-06-30), pages 4 - 9 *
沈志东 等: ""基于分布式扰动的文本对抗训练方法"", 《计算机工程》, vol. 49, no. 9, 30 September 2023 (2023-09-30) *

Also Published As

Publication number Publication date
CN117541367B (en) 2024-04-02

Similar Documents

Publication Publication Date Title
US11574077B2 (en) Systems and methods for removing identifiable information
Maleki et al. A comprehensive literature review of the rank reversal phenomenon in the analytic hierarchy process
CN110797124A (en) Model multi-terminal collaborative training method, medical risk prediction method and device
AU2018202830A1 (en) Digital Asset Platform
CN106537333A (en) Systems and methods for a database of software artifacts
US20230281238A1 (en) Facial test database management system for detection of facial recognition device, and method
CN103455589B (en) Product data moving method under product factory mode, Apparatus and system
JP2014137722A (en) Rule generation device and extraction device
WO2021217987A1 (en) Text summary generation method and apparatus, and computer device and readable storage medium
Xu et al. CET-4 score analysis based on data mining technology
CN117541367B (en) Digital bidding document making and evaluating system based on artificial intelligence
Alitto et al. PRODIGE: PRediction models in prOstate cancer for personalized meDIcine challenGE
Gordon et al. Comparing requirements from multiple jurisdictions
KR101456189B1 (en) Method for evaluating patents using engine and evaluation server
KR101658890B1 (en) Method for online evaluating patents
CN109190102A (en) The system and method that project of inviting outside investment negotiation scheme automatically generates
Lv et al. Counterfactual cross-modality reasoning for weakly supervised video moment localization
Ahmed et al. Software architecture of a learning apprentice system in medical billing
Tun et al. Federated learning with diffusion models for privacy-sensitive vision tasks
Barth et al. Detecting Stale Data in Wikipedia Infoboxes.
Koscinski et al. On-demand security requirements synthesis with relational generative adversarial networks
CN117493646B (en) Intelligent library borrowing tracking system based on blockchain technology
Vero et al. Programmable Synthetic Tabular Data Generation
CN105354201B (en) The method and system screened and eliminate false positive results
Zhang et al. Technical evolution and prediction of blockchain based on different evolution patterns by text mining and bibliometric methods

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant